PRDM1 Drives a TIM3+ Macrophage Immunosuppressive Niche via LGALS9 Signaling in Prostate Cancer Progression

Prostate cancer (PCa) responds poorly to immunotherapy. We investigated the myeloid checkpoint TIM3 (HAVCR2) to define its lineage localization and regulatory logic in the PCa microenvironment.

We integrated stage-resolved public single-cell RNA-seq datasets spanning primary PCa, metastatic hormone-sensitive PCa, and castration-resistant PCa. Myeloid compartments were analyzed via differential expression, regulon inference, and ligand–receptor modeling. Clinical relevance was evaluated in the Cancer Genome Atlas prostate adenocarcinoma (TCGA-PRAD) cohort and independent cohorts using a myeloid TIM3 signature. Mechanistic validation was achieved through PR domain zinc finger protein 1 (PRDM1) chromatin immunoprecipitation followed by Chromatin Immunoprecipitation (ChIP)–qPCR (ChIP-qPCR), TIM3-promoter luciferase assays, and functional perturbation of the galectin 9 (LGALS9)-TIM3 signaling pathway in macrophages differentiated.

TIM3 expression was predominantly confined to monocytes/macrophages, indicating TIM3 as a microenvironmental checkpoint in PCa. TIM3_high macrophages formed a Secreted Phosphoprotein 1 (SPP1)-enriched tumor-associated macrophage (TAM) state coupled to chemokine programs and extracellular matrix remodeling. Regulon profiling nominated PR domain zinc finger protein 1 (PRDM1) as an upstream driver; PRDM1 correlated with TIM3. Communication inference further highlighted an LGALS9-TIM3 axis and a C-X-C motif chemokine receptor 4/integrin subunit beta 1 (CXCR4/ITGB1)-associated permissive niche. Recombinant LGALS9 induced TIM3-linked M2-like macrophages polarization and increased CXCR4/ITGB1, which was attenuated by TIM3 blockade.

Our results delineate a PRDM1-licensed TIM3_high macrophage program sustained by an LGALS9-TIM3 reinforcement loop and coupled to immunosuppressive and remodeling-associated phenotypes. Targeting TIM3 in the myeloid compartment, alone or in rational combinations, may represent a feasible strategy to reprogram tumor-associated macrophage states in PCa.